Shim stack arrangement source material 3

[mburgess]

I am looking to develop a mathmatical model on shim stack layouts. Not to try and simulate the dampening curves, but to be able to compare the differences between changes.
Does anyone know of any source material (papers, books, etc) that may assist.
Regards,
MB

 

[MoreWing]

I know I've looked for the same thing, but not had much success. There is one thing I've found.

http://www.shimprogram.com/index.php

This guy, Kevin, has written a program to compare relative stiffnesses of 2 different shim stacks. It's interesting, but not really in depth enough for me to pay for it.

I'd love to have a program that could give me a force vs. displacement graph of a shim stack. Once you get that, you can apply some flow coefficients and do some sort of damping curves. I think Kevin could probably get there if he wanted to, he seems pretty clever.

Having said this, I've never ran across any paper, book, or site that breaks it all down. Because of this, I end up shooting in the dark and running it on the dyno to see, just like everyone else. It ain't pretty, but once you've done it a couple thousand times, you kind of start to figure out what it takes to get where.

See what you think about the program sample and write back. This might be just what you're looking for.

 

[mburgess]

MoreWing,
The shimprogram.com is fairly close to what I am trying to achive. From what I have read on various forums that the author has been working on the program for about 12 years. So I gather that it would be a large undertaking. Like yourself I have been using the suck it & see approach. The idea of trying to generate a model was to try and gain a greater understanding of what is happening when a shim stack deflects. I was thinking using a simulation based on leaf springs may be a place to start.

 

[SusTestEng]

I remember one supplier saying something like the stiffness of the stack of a particular diameter is proportionate to the sum of the thicknesses squared. So lets say you have 1 shim that is 0.4 thick, and you want to have the same stiffness by using 0.2 thickness shims. If what I remember is correct, it would mean you need four 0.2 thickness shims to equal 1 0.4 thickness shim:

0.4^2 = 0.16
0.2^2 + 0.2^2 + 0.2^2 + 0.2^2 = 0.16

These two may give the same D/F, But I can tell you the feel EXTREMELY different.

I think this is fairly true for comparing the same diameter shims, but when you start stacking different diameters, I have no idea.

You realize that a program that does this is not all that usefull, unless you are making shocks to order in an aftermarket setting. Because damper tuning is all about feel and very little to do with numbers. Everyone always wants to know the damping force in a certain vehicle, but I could care less. I can make 30 different dampers with teh same exact damping force, and each feels completely different. Untill someone can write a program that incorporates friction between each shim, fluid dinamics, response, and the calabration of my backside, a computer program is not worth the time.

 

[MoreWing]

Hey STE,

I'm going to both agree and disagree with you.

I'm fairly sure the Ohlins manual says that it takes something like 7 0.15mm shims to equal the stiffness of 1 0.30mm shim. I've slept since I read that book, so all bets are off.

Getting a certain damper curve is a highly subjective thing to do. There are many avenues to get essentially the same force / velocity curve. I actually think that we measure them incorrectly. We should make make 3 dimensional graphs with frequency as the 3rd axis. That might give us a little more insight to why a damper acts the way it does. I completely agree with your statement that 2 shocks with the same damper curve can feel completely different. I've seen it many times, and it never fails to amaze me. Humans can really be quite sensitve little creatures.

When you are building a shock for a racecar or bike, you generally have 1 person doing it. That person will have their pet theory on how they go about getting a certain curve. Once they get that curve, they may want to add compression or add a little rebound or whatever to fix whatever ill they are fighting. Having a program where you could stick in the initial shim stack, and then stick in a slightly altered shim stack (usually you're only changing a shim or 2) would be very useful even if it didn't incorporate all the intricacies of a real damper. Since all most of us really have to quantify our changes is a crank pin dyno, that's basically how we have to approach the issue, whether it's right or not. Of course, the proof is always in the puddin', and we won't know if we've really made progress until we drive it.

 

[NormPeterson]

The plate constant ("D" in Roark's book) that finds its way into the formulas for flat plate deflections and rotations is a 3rd power function of plate thickness. Hence a 2:1 thickness ratio would correspond to 1:8 for the number of plates/shims required if there are no other effects present. So the conversion of 7 thin plates to equal one plate of twice the thin plate thickness at least sounds reasonable.

Norm

 

[SusTestEng]

Maybe it was the sum of cubes then. I knew if I brought up the idea, someone would correct me or agree.

 

[mburgess]

Morewing,
you have brought up another matter. Is it 'better' (I know alot of suspension tuning is subjective to the driver/rider) to use a less amount of thicker shims or a larger amount of thin ones. As you stated you would possibly need a frequency response axis. Thinking about this I could see the possibility of friction between alot of shims dampening their own response, how much of this is a real effect is another story. Again the whole exercise is wholy dependent on the piston used. I find that there is alot of mis information about.
MB

 

[SusTestEng]

mburgess,

Most people prefer the "feel" of more/thinner shims to few/thick shims. I think the reason is that the friction between the shims causes a smoother transition, and makes the damping force feel less "digital". The only time I use a thick shim vs. a stack of thin ones is when a very high response is needed, like if bottoming becomes a problem, or there is just no more room on the bolt to have a stack of shims that tall.

 

[MoreWing]

In back to back tests, I've always leaned towards the big stack of thin shims. Drivers seem to use the words, 'progressive' and 'smooth' when giving feedback. I think that STE's 'digital' comment is spot on.

I don't try to prop the car up with the shocks, that's just not how I approach it. Some people do, so that probably means they go with the thicker shim approach. This might be the ticket if the rest of the car is tuned around it. Like I said, everyone has their pet theories.

I think that the feel of the thin shims has something to do with the friction in the stack, but I think it also might have something to do with the initial displacement of the shim stack, which happens easier than if you the single thick sealing shim.

Using a bunch of thin shims is not without it's problems. I've had trouble with 0.1mm shims yielding and also with fatigue issues. In general, they require more maintanance. It's also more difficult to get shocks to match from side to side on the car because a small manufacturing tolerance on each shim can have quite a stackup if you are using 10-15 shims. Sometimes I end up throwing out the whole stack and sticking a new one in. More often than not, that miraculously fixes it. Go figure.

 

[NormPeterson]

Sounds like we're talking about damping of the moving parts within the damper . . . maybe that also goes toward explaining the Ohlins' 7:1 vs the theoretical 8:1? Does that make any sense?

Norm

 

[MoreWing]

I just grabbed my Ohlins manual. It looks like the problem was that I was mis-quoting them. Hey, I said I had slept since I last read the book, give me a break.

Here's the example it gives. How many 0.15mm shims to make the same force as a single 0.30mm shim?

(0.30 / 0.15)^3 = 8

It takes 8 0.15mm shims to have the same effect as 1 0.30mm shims.

It then adds the caveat, "Because the shims do not bend exactly like constant section beams, this formula gives an answer that is not entirely correct. In reality, the equivalency factor is slightly more that the result yielded by the equation."

So, there you go.

 

[mburgess]

Ok this is interesting as I had been going down the path of using fewer but thicker shims. The application was in the front forks of a road racing motorcycle. However this was in conjunction of using a piston with a smaller port area, the idea being to place more point force on the shim stack for greater deflection of the stack than some of the big port designs.